Microsoft today announced that it:
is supporting the addition of Microsoft's exFAT technology to the Linux kernel.
Microsoft has published the exFAT file system specification on its Windows Dev Center site.
While the code remains under copyright, Microsoft also stated that the exFAT code incorporated into the Linux kernel will be available under GPLv2.
We also support the eventual inclusion of a Linux kernel with exFAT support in a future revision of the Open Invention Network's Linux System Definition, where, once accepted, the code will benefit from the defensive patent commitments of OIN's 3040+ members and licensees.
It is noteworthy that there is already a free and open source exFAT driver available for FreeBSD and multiple Linux distributions, but it is not an official part of the Linux kernel due to the patent encumbrance of exFAT.
Also at TechCrunch and VentureBeat.
(Score: 4, Insightful) by PartTimeZombie on Thursday August 29 2019, @03:57AM (17 children)
I wonder what Microsoft's end game is here. I assume its the old embrace, extend, extinguish like they have done in the past, but I wonder if they are powerful enough to do that anymore?
(Score: 2, Informative) by Anonymous Coward on Thursday August 29 2019, @05:10AM (1 child)
A quick check on Wikipedia says that exFAT was introduced in 2006, so it has at most 2 years of patent protection in the United States. Sounds like they are just buying some goodwill and maybe longevity for a product nearing the end of its usefulness.
(Score: 1, Informative) by Anonymous Coward on Thursday August 29 2019, @02:14PM
I don't understand how you come to this 2 year number. Current United States patent law is that patents last 20 years from the first filing date.
In principle patents can be filed up to 1 year after public disclosure of an invention, so the presumed date where all relevant exfat patents should be expired (at least for the original implementation) is ca. 2027.
(Score: 5, Insightful) by canopic jug on Thursday August 29 2019, @05:38AM (12 children)
The end game is the same as before: EEE.
Notice that M$ has avoided releasing any of the patents, and that the format itself is useless in a modern context. That tells more than anything else about what their goals are. The only thing that ends patent liability is time. M$ keeps changing and patenting the FAT series to keep the patents active. Unlike with trademarks, it is perfectly legal to let patented technology spread until people are dependent on it and then, only then, demand full payment at a price of their own choosing.
Besides, exfat is possiby the worst, least appropriate file format in existence. Its very design is so bad it is almost as if it is intended to lose data, especially with large files. If M$ were serious about showing any change in regards to its treatment of FOSS, one of the things it could and should do would be to roll out FULL support for EXT4 or maybe also OpenZFS for its whole product line and then push as hard as it did for exfat to get those better formats into the market.
tldr; submarine patents [moneyterms.co.uk]
Money is not free speech. Elections should not be auctions.
(Score: 2, Interesting) by Anonymous Coward on Thursday August 29 2019, @06:04AM (2 children)
This is the right answer. ExFAT is terrible. If MS actually wants interoperability it should go both ways, and EXT4 should be treated as a first class citizen in Windows.
I'm not holding my breath.
(Score: 2) by takyon on Thursday August 29 2019, @06:13AM (1 child)
2-128 TB SDUC and microSDUC use exFAT by default. So it's probably a good thing to have that support ready.
[SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]
(Score: 3, Informative) by Anonymous Coward on Thursday August 29 2019, @06:57AM
In order to certify that your product is SDXC (finalized circa 2009) compatible you have to support exFAT in a default configuration. That is why many SBCs, especially the bigger targets, don't advertise their support for that format, despite being electrically compatible. They have to reformat the card with another file system, and thus violate the specification.
Note also that vanilla/HC/XC/UC capacity distinction is different than having the UHS bus available because cards are supposed to be backwards compatible to the old bus at slower speeds.
(Score: 5, Informative) by c0lo on Thursday August 29 2019, @07:01AM (8 children)
Almost, but not quite here, nor there.
From TFMicrosoftAnnouncement link, with my emphasis:
Just a vague promise that "will eventually include the exFAT in the Open Invention Network at an unspecified time in the future, so that you'll not have to worry about them patents".
The only rational reaction that I see is: "You first release it in Open Invention Network and only then we'll include it into the Linux Kernel"
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 4, Interesting) by canopic jug on Thursday August 29 2019, @10:02AM (7 children)
The only rational reaction that I see is: "You first release it in Open Invention Network and only then we'll include it into the Linux Kernel"
That's unlikely given current and past activities. M$ is going to continue holding on to those software patents until they expire. Here's what one of the LWN participants with relevant domain expertise had to say:
It's also likely that they'll just make a slight modification and refile at the USPTO to get another 17+ years for nothing. From a performance perspective, as well as for other technical reasons, the exFAT file system is garbage and needs to go. Using exFAT as a vehicle to injevt M$ software patents into the kernel isn't showing that the company has any intention of reforming. Once in the kernel, M$ has the legal right, in the US, to require payment from anyone using Linux. Such are patents.
If M$ wanted to show support for Free and Open Source Software, it would actually support it. In this discussion, adding EXT4 support would do that. Or even EXT2 support would be a good move. Both are worlds better than exFAT. Or another area of improvement would be to actually support the Open Document Format instead of continuing to break it and undermine its use. Neither will happen any time soon if at all because that would cede control and this fight is about control of the computers more than it is about money.
Money is not free speech. Elections should not be auctions.
(Score: 0) by Anonymous Coward on Thursday August 29 2019, @01:55PM
I think this is prompted by the fact that they are making much of their money off the SAS Azure system.
Many of the VMs in that system are running Linux, and customers are asking for better integration. They seem to recognize that listening to customers might actually be a good idea. It helps that they are competing with Amazon in this space.
(Score: 2) by Pino P on Thursday August 29 2019, @03:34PM (3 children)
Threatening a patent infringement suit against users of Linux would appear to cost Microsoft the right to distribute Linux to its Azure subscribers:
(Score: 3, Touché) by hendrikboom on Thursday August 29 2019, @06:14PM (1 child)
Would the patent interfere with distribution or with use?
(Score: 0) by Anonymous Coward on Thursday August 29 2019, @06:54PM
Both. That section of the GPL gives two options to distributors. They can either stop using it all together, or rip out that modification from the GPL program. Even though Microsoft can grant the use to their direct customers, they have to also grant it to everybody those customers could give it (which ends up meaning everyone on the planet due to GPL's viral nature) in order to comply with the GPL.
As for users, if they get and use a kernel with exFAT in it without getting the patent license, then they are directly infringing the patent and can't use the software anyway. If they use the kernel with a patent license, but cannot transfer or sublicense that licence to everyone, then you are in violation of Section 7 of the GPL and cannot use said software at all.
(Score: 2) by PinkyGigglebrain on Friday August 30 2019, @07:00PM
As if that would stop them. This is Microsoft we are talking about, they have the legal and financial resources to drag any court case arising from their violating the GPL on for decades.
And lets be honest, there aren't any GNU/Linux or GPL supporting organizations that have the resources to match Microsoft's in a long fight.
"Beware those who would deny you Knowledge, For in their hearts they dream themselves your Master."
(Score: 2) by bzipitidoo on Thursday August 29 2019, @09:56PM (1 child)
> the exFAT file system is garbage and needs to go.
This. A long time ago, I tested the performance of a bunch of file systems. There were only slight variations between ext[234], xfs, zfs, btrfs, and a few others. But FAT (FAT32) stood out as much, much slower than all the rest. exFAT supposedly has some performance improvements.
Plain copying of critical data, as FAT does by having a duplicate File Allocation Table, is the most brain dead simple, wasteful, and ineffective way to guard against errors in the data. Any interruption during a write, such as by a power failure or removal of the medium, can unrecoverably corrupt a FAT file system more easily than just about any other file system. Then there's the hard size limits. From the 8.3 limit on the names from way back in the day, to the FAT32 4G file size limit and 65K directory entry limit, FAT's limits were always too low, requiring frequent modification as technology exceeded FAT's capacities again and again.
I have an all-in-one that can write scans to flash drives-- as long as the flash drive is formatted with FAT. But, must be careful. If there is not enough free space left on the flash drive, the device will just keep on writing, paving over critical file system structures, completely borking it.
If those technical deficiencies aren't enough to get FAT permanently discontinued, there's M$'s long history of leveraging patents on FAT to extort money from others. It's possible FAT's limits were purposely calculated to give M$ the excuse it needed to roll out a few fresh patents on the latest meager expansion of yet another limit. FAT would be dead and gone if it wasn't the default on flash drives, and if M$ Windows supported more than just FAT and NTFS. If we could format a flash drive with ext2, and it would just work in Windows, that would eliminate one of the few reasons left to use FAT. Possibly that would leave thee embedded world as the last holdout.
(Score: 2) by c0lo on Friday August 30 2019, @12:09AM
From the 8.3 limit on the names from way back in the day, to the FAT32 4G file size limit and 65K directory entry limit, FAT's limits were always too low, requiring frequent modification as technology exceeded FAT's capacities again and again.
Oh, come on! Are you now gonna tell me 640K is not enough for everyone?
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 2) by rylyeh on Friday August 30 2019, @02:29AM (1 child)
After spedning untold amounts of $ on code that has to be mostly throw away every 5-10 years, I think they are seeing the writing on the wall.
Open source is better. Especially now that they are branding themselves a 'service'.
"a vast crenulate shell wherein rode the grey and awful form of primal Nodens, Lord of the Great Abyss."
(Score: 2) by PartTimeZombie on Friday August 30 2019, @02:47AM
I wonder what Microsoft will be in 10 years time?